W1
An AI coding agent that lives in your editor. Describe what you want — a feature, a fix, a whole app — and W1 builds it in your project: editing files, running commands, and checking its own work, live in the side panel.
No API keys. No setup. W1 runs on a hosted backend, so you just sign in and start building — nothing to configure, no provider keys on your machine.
🔑 Closed beta
W1 is invite-only right now. You'll need a beta key to sign in — join the waitlist at w1lab.com and we'll send you one.
What W1 does
- Builds — "add Google sign-in", "make a landing page for my app", "scaffold a REST API."
- Fixes & refactors — "find why checkout fails and fix it", "split this 800-line file into modules."
- Explains — ask how part of your codebase works; W1 reads it and tells you.
- Designs from images — paste or drop a screenshot and W1 builds to match.
Every run happens in your real project, and you watch each step as it goes.
Get started
- Install W1 (you're here) and reload your editor.
- Click the W1 icon in the sidebar → Sign in.
- Paste your beta key when prompted. (No key yet? Join the waitlist.)
- Open a folder, type what you want, and hit send.
💡 Have Node and git installed for best results — W1 uses them to run and verify your changes — but it works without them.
Why W1
- Just sign in — no keys, no model setup, no per-token billing to wire up.
- It checks its own work — runs your tests, lints, and builds, and reports the real results.
- Skills on demand — W1 pulls in specialized abilities (frontend design, documents, testing…) only when a task needs them.
- Stays out of your repo — W1 keeps its working files to itself; your codebase stays clean.
Privacy
W1 runs on a hosted backend, so model calls and a short record of each run leave your machine:
- Run traces (on by default) — a trimmed record of each run (your prompt, the plan, command results, and a code diff capped at 20,000 characters) is sent to W1 to diagnose and improve the agent. No keys or secrets are included. Turn it off with the
w1.shareTraces setting.
- Usage analytics — anonymous product stats (runs, tokens, latency, errors). No source code is sent.
- Set VS Code's
telemetry.telemetryLevel to off to disable both.
Full policy: app.w1lab.com/privacy
Settings
| Setting |
What it does |
w1.shareTraces |
Share run traces to help improve W1 (default on). See Privacy. |
w1.voice |
Narrate what W1 is doing (default on). |
Made for builders. Join the waitlist →